Growing global concerns about sustainability have pushed organizations to prioritize social innovation, which not only helps solve social problems but also offers competitive advantages. However, social innovation presents resource challenges. Many organizations are turning to big data analytics capabilities to address these challenges, as they are capable of facilitating organizations to address social issues and create sustainable values. Despite the potential, the impact of big data analytics capabilities on social innovation remains under-researched. Most studies focus on economic performance rather than social impact. This study aims to address this gap by exploring the relationship between big data analytics capabilities and social innovation using the organizational information processing theory and organizational learning theory. The study hypothesizes that big data analytics capabilities enhance social innovation through knowledge management, specifically knowledge exploration and exploitation. It further investigates the joint impact of knowledge exploration and exploitation on social innovation, anticipating a synergistic effect when both are high and a decline in social innovation when there is an imbalance between the two.

The organizational information processing theory suggests that organizations facing uncertainty need enhanced information processing capabilities to achieve optimal performance. Big data analytics, as an advanced information system, improves this capability, facilitating innovation. The organizational learning theory emphasizes the importance of knowledge exploration (seeking new knowledge) and exploitation (applying existing knowledge) for organizational competitiveness and innovation. Existing studies frequently link big data analytics capabilities with various types of innovation, primarily from a dynamic capability or resource-based view. However, research on the relationship between big data analytics capabilities and social innovation in for-profit organizations remains limited. This study defines social innovation as a process solving difficult social problems to create social and economic value. Antecedents of social innovation explored in previous research include organizational factors (social entrepreneurship, knowledge networks, corporate strategic orientation), social factors (institutional and environmental factors), and technological factors (IT, big data analytics). Organizational ambidexterity theory highlights the benefits of simultaneous exploration and exploitation of knowledge for competitive advantage and sustainability. IT capabilities are crucial for achieving this knowledge ambidexterity.

This study employed a quantitative research method using a survey of 354 Chinese high-tech firms. The sample was randomly selected from high-tech companies in Beijing, Zhejiang, Jiangsu, and Guangdong, known for high information technology adoption and a focus on socially responsible technology use. The survey used a two-stage approach: a questionnaire to CIOs in Time 1 (T1) measuring big data analytics capabilities and knowledge ambidexterity, followed by a questionnaire to CEOs in Time 2 (T2) measuring social innovation. The questionnaires were adapted from established scales and validated by experts. The study used seven-point Likert scales for all items. Big data analytics capabilities were measured using three dimensions: management, technology, and personnel capabilities. Knowledge exploration and knowledge exploitation were measured using adapted scales from previous research. Social innovation was the dependent variable, measured using a five-item scale. Firm age, size, and industry were included as control variables. Data analysis involved reliability and validity tests (Cronbach's alpha, KMO, Bartlett's test, AVE, CR), common method bias assessment (Harman's one-factor test), correlation analysis, structural equation modeling (SEM) using AMOS, and bootstrap mediation analysis to test the mediating roles of knowledge exploration and exploitation. Polynomial regression and response surface analysis (using SPSS) were used to examine the joint effects of knowledge exploration and exploitation on social innovation.

The study found significant positive relationships between big data analytics capabilities (management, technology, and personnel) and social innovation. All three dimensions of big data analytics capabilities significantly and positively predicted social innovation (H1a, H1b, H1c supported). Big data analytics capabilities also significantly and positively influenced both knowledge exploration (H2a, H2b, H2c supported) and knowledge exploitation (H3a, H3b, H3c supported). Both knowledge exploration and knowledge exploitation individually had significant positive relationships with social innovation (H4a and H4b supported). Bootstrap mediation analysis showed that both knowledge exploration and exploitation fully mediated the relationship between big data analytics capabilities and social innovation (H5a-c and H6a-c supported). Polynomial regression and response surface analysis indicated that social innovation was highest when both knowledge exploration and exploitation were high (H7a supported) and decreased when there was an imbalance between the two (H7b supported). The response surface analysis showed the optimal configuration of knowledge exploration and exploitation was a state of balance, where both are high.

The findings support the hypotheses that big data analytics capabilities positively affect social innovation, and that this effect is mediated by knowledge exploration and exploitation. The results highlight the importance of balanced knowledge management strategies for maximizing social innovation. The study provides empirical evidence for the organizational information processing theory and organizational learning theory in the context of social innovation. The mediating role of knowledge ambidexterity underscores the need for organizations to effectively manage both the exploration of new knowledge and the exploitation of existing knowledge. The study's findings contribute to the growing body of research on big data analytics and innovation, specifically extending it to the domain of social innovation.

This study provides valuable insights into how big data analytics capabilities can foster social innovation, emphasizing the crucial mediating role of knowledge ambidexterity. The optimal configuration of high and balanced knowledge exploration and exploitation is key. Future research could explore cross-cultural comparisons, investigate other mediating variables, and utilize objective data for enhanced validity.

The study's findings are limited to Chinese high-tech firms. The use of questionnaires introduces subjective bias. Future research should consider cross-cultural validation, explore additional mediating factors, and utilize objective data for stronger generalizability.